Standardized interviewing of patients or research subjects using computers has many potential benefits to both biomedical research and clinical practice in terms of acceptability, reliability, accuracy, quality of data and ultimately cost-efficiency. Computer-assisted self-interview with audio (A-CASI) systems have become the gold standard in the social survey (particularly substance abuse) research field but have not gained a widespread acceptance in other settings. This is due in part to the fact that the development of computerized instruments (unless very simple in structure or content) is a lengthy and difficult process that requires skilled programmers or the mastery of complex scripting languages within expensive survey software. [unreadable] [unreadable] The overall goal of this research is to develop an innovative software tool (CASIA) that will enable researchers and clinicians to rapidly and efficiently create their own automated versions of any standardized measure without the need for programmers or the need to learn a programming language. The specific aims of this proposal are to: [unreadable] [unreadable] (1) Investigate the usability of CASIA, an authoring tool for use by non-programmers, in the conversion of standard interviews and questionnaires into computer-assisted personal interview (CAPI) and computer- assisted self-interview with audio (A-CASI) systems. [unreadable] [unreadable] (2) Develop a cognitive framework of computer-assisted interview authoring that describes the cognitive processes and knowledge pertaining to the creation of automated versions of questionnaires, rating scales and structured interviews. [unreadable] [unreadable] (3) Utilize this framework to inform a user-centered design approach to revise and extend the CASIA system thereby providing a set of resources in the authoring environment that can adapt to the characteristics of users, tasks and contexts. [unreadable] [unreadable] The enhanced usability of this authoring-tool, the result of a cognitively-informed and user-centered design approach, will make it more likely that A-CASI systems will be developed, achieve a broad usage and, most importantly, persist in settings long-term. [unreadable] [unreadable] [unreadable]